Commercial switchable glazing devices, also commonly known as smart windows and electrochromic window devices, are well known for use as mirrors in motor vehicles, aircraft window assemblies, sunroofs, skylights, and architectural windows. Such devices may comprise, for example, active inorganic electrochromic layers, organic electrochromic layers, inorganic ion-conducting layers, organic ion-conducting layers and hybrids of these sandwiched between two conducting layers. When a voltage is applied across these conducting layers the optical properties of a layer or layers in between change. Such optical property changes typically include a modulation of the transmissivity of the visible or the solar sub-portion of the electromagnetic spectrum. For convenience, the two optical states will be referred to as a bleached state and a darkened state in the following discussion, but it should be understood that these are merely examples and relative terms (i.e., a first one of the two states is more transmissive or “more bleached” than the other state and the other of the two states is less transmissive or “more darkened” than the first state) and that there could be a set of bleached and darkened states between the most transmissive state and the least transmissive state that are attainable for a specific electrochromic device; for example, it is feasible to switch between intermediate bleached and darkened states in such a set.
The broad adoption of electrochromic window devices in the construction and automotive industries will require a ready supply of low cost, aesthetically appealing, durable products in large area formats. Electrochromic window devices based on metal oxides represent the most promising technology for these needs. Typically, such devices comprise two electrochromic materials (a cathode and an anode) separated by an ion-conducting film and sandwiched between two transparent conducting oxide (TCO) layers. In operation, a voltage is applied across the device that causes current to flow in the external circuit, oxidation and reduction of the electrode materials and, to maintain charge balance, mobile cations to enter or leave the electrodes. This facile electrochemical process causes the window to reversibly change from a more bleached (e.g., a relatively greater optical transmissivity) to a more darkened state (e.g., a relatively lesser optical transmissivity).
TCO materials typically used in electrochromic windows such as FTO and ITO react with lithium at voltages below ˜1V vs. Li/Li+, lowering their electrical performance and darkening the material. Electrolytes typically incorporated into the ion conductor, or the presence of water or protic impurities, have voltage stability windows between ˜1 and ˜4.5 V vs. Li/Li+. Therefore, it is beneficial to use electrode materials that undergo redox events within these limits. For example, tungsten oxide (WO3) is a well known cathodic electrochromic material that is bleached at ˜3.2 V vs. Li/Li+ and darkens upon reduction, typically to ˜2.3 V vs. Li/Li+. Consequently, electrochromic devices comprising a tungsten oxide cathode are common.
Certain nickel oxide and hydroxide based materials darken anodically to produce a darkened state transmission spectrum that is complementary to lithiated WO3 and it is a popular target to partner WO3 in electrochromic windows. Certain methods for the preparation of lithium nickel oxide films (LiNiOx) have been reported in the literature. These include sputter methods (see, e.g., Rubin et. al. Solar Energy Materials and Solar Cells 54; 998 59-66) and solution methods (see, e.g., Svegl et. al., Solar Energy V 68, 6, 523-540, 2000). In both cases the films exhibit high area charge capacity (>20 mC/cm2), with bleached state voltages of ˜1.5V. This bleached state voltage is relatively close to the reaction potential of lithium with typical TCO materials, the lower voltage limit of common electrolytes and the reaction potential required to over-reduce lithiated nickel oxides to nickel metal, a cathodic electrochromic reaction. The proximity of the bleached state voltage to such degrading mechanisms presents significant device control issues: methods will be required to consistently drive the device to the bleached state without driving the anode into damaging voltage regimes accommodating, for example, issues such as local electrode inhomogeneity. Furthermore, the bleached state lithiated nickel oxide cannot typically be handled in air without the material performance degrading. The lack of air stability of the bleached state of un-doped lithium nickel oxide films is demonstrated in the examples section of this invention where lithium nickel oxide films were prepared using liquid mixtures of lithium and nickel salts to produce, after thermal processing, films in their darkened state. Upon electrochemical or chemical reduction to their lithiated forms in an inert atmosphere bleached state films were produced but upon exposure to air in this bleached state they quickly lose the reversible electrochromic properties.
Examples of sputter coated lithiated nickel oxides that contain a second metal have been reported. For example, U.S. Pat. No. 6,859,297 B2 describes the lithiation (and bleaching) of mixed nickel oxide films that contain Ta and W in appreciable quantities. The material was prepared by a two step process, the first step being a vacuum co-sputtering process to produce a mixed Ta/Ni oxide film and second electrochemical lithiation step to produce a material in its bleached state. The Ta-containing oxide films are characterized extensively and have no long range order of evidence of crystallinity by XRD and, required handling in a controlled atmosphere to preclude their exposure to water and oxygen.
A wide range of structures derive from metal occupation of the octahedral and tetrahedral sites within close packed anion arrays. In such arrays, there are equal numbers of octahedral sites as anions and twice as many tetrahedral sites as anions. The term “rock salt” as used herein describes a cubic structure in which metal cations (“M”) occupy all of the octahedral sites within a close packed anion array, resulting in the stoichiometry MO. Furthermore, the metals are indistinguishable from one another regardless of whether the metals are the same element or a random distribution of different elements. In the specific case of NiO, for example, the cubic rock salt unit cell has a ˜4.2 Å and a largest d-spacing of ˜2.4 Å. In the case where there is more than one type of metal, different structures are created depending upon how and if the metals order themselves over the octahedral and tetrahedral holes. The case of LixNi1-xO is instructive: for all values of x, the oxygen anions are close packed and the metals are arranged on the octahedral sites. For values of x less than ˜0.3, the lithium and nickel cations are randomly arranged; for values of x greater than 0.3, the metals segregate to create nickel-rich and lithium-rich layers, creating layered structures with hexagonal symmetry. The end member, Li1/2Ni1/2O (equivalently, LiNiO2) is formed from alternate layers of —Ni—O—Li—O— with a hexagonal unit cell (a=2.9, c=14.2 Å) and a largest d-spacing of ˜4.7 Å. The voltage associated with the lithium intercalation events is above 3V vs. Li/Li+.
Even though all of the octahedral sites in LiNiO2 are full, additional lithium can be inserted into the material, forming Li1+xNiO2. The additional lithium necessarily occupies sites in close proximity to other cations with less shielding from the anion array. Thus, the insertion of this additional lithium occurs at lower voltages, <2V vs. Li/Li+ for bulk phase materials.
Other phases that are possible from metal occupation of sites within close-packed oxygen arrays include the orthorhombic phases Li1/2Ni1/3Ta1/6O and Li1/2Ni1/3Nb1/6O in which the Nb or Ta segregate to one set of octahedral sites and the Ni and Li are mixed on the remaining sites. Further examples are the spinel phases including Li1/4Mn3/8Ni1/8O in which Mn and Ni occupy the octahedral sites and Li occupies ¼ of the tetrahedral sites.
A collective signature of all of the phases described above are the close packed layers. In the rock salt structure, these give rise to a single diffraction reflection at ˜2.4 Å, labeled as the (111) reflection. This is the largest symmetry allowed d-spacing in the rock salt structure. The second largest d-spacing allowed in the rock salt structure is the (200) peak whose d-spacing is ˜2.1 Å. In lower symmetry structures such as Li1/2Ni1/2O and Li1/2Ni1/3Ta1/6O, reflections equivalent to the rock salt(111) and (200) reflections are observed at approximately the same d-spacing but are labeled differently and may be split into multiple peaks. For example, in the hexagonal, layered material the rock salt (111) reflection splits into two reflections, the (006) and the (102) peak, both of which occur at ˜2.4 Å and the rock salt (200) peak becomes the (104) peak, whose d-spacing is also 2.1 Å. A clear signature that an ordered metal sub-lattice exists within a material giving rise to structures such as Li1/2Ni1/2O, Li1/2Ni1/3Nb1/6O, and Li1/4Mn3/8Ni1/8O is the presence of reflections with d-spacings greater than 2.4 Å (Table 1).
TABLE 1Largest d-spacing (Å) and associated hkl of examplematerials derived from metals within octahedral and/ortetrahedral sites created by close packed oxygen arraysLargestCompositionStructure Noted-spacing (Å)hklNiOrock salt2.4(111)Li0.1Ni0.9Orock salt, Li and Ni randomly2.4(111)arrangedLi1/2Ni1/2OHexagonal, Li and Ni ordered4.7(003)into layersLi1/2Ni1/3Ta1/6OOrthorhombic, Ta and Li/Ni4.7(111)orderedLi1/4Mn3/8Ni1/8OCubic, Ni/Mn in octahedral4.7(111)sites; Li in tetrahedral sites
Although a range of electrochromic anodic materials have been proposed date, there is a need for anode films that can be prepared by simple single-step deposition processes to produce EC anodes with improved thermal stability, high optical clarity in their as-deposited states, and that can be tuned via composition and film thickness to adopt a wide variety of area charge capacities and optical switching properties.